The GAPS Programme with HARPS-N at TNG - The WASP-33 system revisited with HARPS-N

Page created by Kenneth Navarro
 
CONTINUE READING
Astronomy & Astrophysics manuscript no. w33_acc                                                                                            ©ESO 2021
                                                May 27, 2021

                                                                        The GAPS Programme with HARPS-N at TNG
                                                                         XXXI. The WASP-33 system revisited with HARPS-N?
                                               F. Borsa1 , A. F. Lanza2 , I. Raspantini3 , M. Rainer4 , L. Fossati5 , M. Brogi6, 7, 8 , M. P. Di Mauro9 , R. Gratton10 , L. Pino4 ,
                                                       S. Benatti11 , A. Bignamini12 , A. S. Bonomo7 , R. Claudi10 , M. Esposito13 , G. Frustagli3, 1 , A. Maggio11 ,
                                                    J. Maldonado11 , L. Mancini14, 15, 7 , G. Micela11 , V. Nascimbeni10 , E. Poretti1, 16 , G. Scandariato2 , D. Sicilia2 ,
                                                A. Sozzetti7 , W. Boschin16, 17, 18 , R. Cosentino16 , E. Covino19 , S. Desidera10 , L. Di Fabrizio16 , A. F. M. Fiorenzano16 ,
                                                                  A. Harutyunyan16 , C. Knapic12 , E. Molinari20 , I. Pagano2 , M. Pedani16 , G. Piotto21
                                                                                                    (Affiliations can be found after the references)
arXiv:2105.12138v1 [astro-ph.EP] 25 May 2021

                                                                                                                  Received ; accepted

                                                                                                                      ABSTRACT

                                               Context. Giant planets in short-period orbits around bright stars represent optimal candidates for atmospheric and dynamical studies of exoplane-
                                               tary systems.
                                               Aims. We analyse four transits of WASP-33b observed with the optical high-resolution HARPS-N spectrograph to confirm its nodal precession,
                                               study its atmosphere and investigate the presence of star-planet interactions.
                                               Methods. We extract the mean line profiles of the spectra by using the Least Square Deconvolution method, and analyse the Doppler shadow and
                                               the radial velocities. We also derive the transmission spectrum of the planet, correcting it for the stellar contamination due to rotation, center-to-
                                               limb variations and pulsations.
                                               Results. We confirm the previously discovered nodal precession of WASP-33b, almost doubling the time coverage of the inclination and projected
                                               spin-orbit angle variation. We find that the projected obliquity reached a minimum in 2011 and use this constraint to derive the geometry of the
                                               system, in particular its obliquity at that epoch ( = 113.99◦ ± 0.22◦ ) and the inclination of the stellar spin axis (is = 90.11◦ ± 0.12◦ ), as well as the
                                               gravitational quadrupole moment of the star J2 = (6.73 ± 0.22) × 10−5 , that we find to be in close agreement with the theoretically predicted value.
                                               Small systematics errors are computed by shifting the date of the minimum projected obliquity. We present detections of Hα and Hβ absorption in
                                               the atmosphere of the planet, with a contrast almost twice smaller than previously detected in the literature. We also find evidence for the presence
                                               of a pre-transit signal, which repeats in all four analysed transits and should thus be related to the planet. The most likely explanation lies in a
                                               possible excitation of a stellar pulsation mode by the presence of the planetary companion.
                                               Conclusions. A future common analysis of all available datasets in the literature will help shedding light on the possibility that the observed
                                               Balmer lines transit depth variations are related to stellar activity and/or pulsation, and to set constraints on the planetary temperature-pressure
                                               structure and thus on the energetics possibly driving atmospheric escape. A complete orbital phase coverage of WASP-33b with high-resolution
                                               spectroscopic (and spectro-polarimetric) observations could help understanding the nature of the pre-transit signal.
                                               Key words. planetary systems – techniques: spectroscopic – techniques: radial velocities – planets and satellites: atmospheres –
                                               stars:individual:WASP-33

                                               1. Introduction                                                                sphere undergoing a significant mass loss, affecting its composi-
                                                                                                                              tion and evolution (e.g., Fossati et al. 2018).
                                               Transiting exoplanets orbiting intermediate mass (e.g., A-type)                     Short-period systems are interesting also because they expe-
                                               stars on short period (Porb ≤5 days) orbits are excellent labo-                rience strong tidal interactions between the planet and its host
                                               ratories for atmospheric and dynamical studies. The high day-                  star. Star-planet tidal interactions could modify the stellar rota-
                                               side equilibrium temperatures of these Ultra-Hot Jupiters (UHJs,               tion rate along the system’s evolution (Gallet et al. 2018), cause
                                               T eq ≥2200 K) cause the thermal dissociation of most molecules                 planet oblateness (Akinsanmi et al. 2019), suppress stellar activ-
                                               (e.g., Lothringer et al. 2018; Arcangeli et al. 2018). Depend-                 ity (Fossati et al. 2018), and/or induce stellar pulsations (de Wit
                                               ing on the heat transfer efficiency in the atmosphere there might              et al. 2017). The measure of the projected spin-orbit angle either
                                               be also variations in the chemical composition of the day-side,                with radial velocities (RVs) through the Rossiter-McLaughlin ef-
                                               night-side and terminator of the planet (e.g., Parmentier et al.               fect (RM, Rossiter 1924; McLaughlin 1924) or with the Doppler
                                               2018). Orbiting at short distances from their parent star, the                 tomography technique (e.g., Collier Cameron et al. 2010) can
                                               strong stellar ultraviolet irradiation makes the planetary atmo-               help to shed light on a system evolution history (e.g., Fabrycky
                                                                                                                              & Winn 2009).
                                               Send offprint requests to: F. Borsa                                                 In this context, WASP-33b (Collier Cameron et al. 2010) is
                                               e-mail: francesco.borsa@inaf.it                                                a very intriguing target. It is a UHJ (Mp ∼2.2 MJup , Rp ∼1.6
                                                 ?
                                                    Based on observations made with the Italian Telescopio Nazionale
                                               Galileo (TNG) operated on the island of La Palma by the Fundacion              RJup ) orbiting in ∼1.21 days a bright δ-Scuti A-type star (V=8.3,
                                               Galileo Galilei of the INAF at the Spanish Observatorio Roque de los           T eff ∼7500 K, v sin i s ∼86 km s−1 ) with a highly misaligned pro-
                                               Muchachos of the IAC in the frame of the program Global Architecture           jected spin-orbit angle λ ∼-110 degrees (Collier Cameron et al.
                                               of the Planetary Systems (GAPS).                                               2010; Lehmann et al. 2015). Stellar non-radial pulsations were
                                                                                                                                                                            Article number, page 1 of 18
A&A proofs: manuscript no. w33_acc

already noted spectroscopically in the discovery paper (Collier       Table 1. WASP-33 HARPS-N observations log.
Cameron et al. 2010). Photometric oscillations of the star were
first reported by Herrero et al. (2011), and the stellar pulsation      Transit number      Night1       Exposure time    Nobs   S/Nave
spectrum has been analysed in different studies (Kovács et al.
                                                                                 1           28 Sep 2016       600s        40     107
2013; von Essen et al. 2014; Mkrtichian 2015; von Essen et
                                                                                2∗           20 Oct 2016       600s        32     90
al. 2020). Because of its easily detected stellar pulsations, the                3           12 Jan 2018       900s        23     166
WASP-33 system has been considered a good candidate for the                      4           02 Jan 2019       600s        33     115
detection of star-planet interactions (Herrero et al. 2011). WASP-      ∗
                                                                          Weather conditions rapidly worsening.
33b was also noted as a possible target for which both classi-          1
                                                                          Start of night civil date.
cal and relativistic node precessional effects could be evidenced
within a reasonable amount of time (Iorio 2011, 2016); indeed
classical node precessional effects have been afterwards detected
by Johnson et al. (2015) and Watanabe et al. (2020).                  and RVs, but only in Sect. 6 where we look at the pre-transit
     Its atmosphere has also been the subject of different studies.   portion of the data.
The probable presence of a temperature inversion in its atmo-
sphere (Haynes et al. 2015; von Essen et al. 2015) was best ex-
plained by the presence of titanium oxide (TiO), whose detection      3. Precession of the orbital plane and stellar spin
with high-resolution spectroscopy is however debated (Nugroho            axis
et al. 2017; Herman et al. 2020). Other results include the first
indication of aluminum oxide (AlO) in an exoplanet by using           WASP-33 is a δ-Sct A-type star. Since no default A-type mask
low resolution spectrophotometry (von Essen et al. 2019), the         is supported by the HARPS-N DRS pipeline (Cosentino et al.
detection of ionized calcium (Ca ii H&K) up to very high upper-       2014), we followed the approach presented in Borsa et al. (2019)
atmosphere layers close to the planetary Roche lobe (Yan et al.       and extracted the mean line profiles by means of the Least
2019) and of Balmer lines (Yan et al. 2021; Cauley et al. 2021).      Square Deconvolution (LSD) software (Donati et al. 1997). This
The existence of a thermal inversion was recently confirmed with      software performs a Least-Squares Deconvolution of the nor-
the detection of Fe i in emission using high-resolution observa-      malised spectra with a theoretical line mask extracted from
tions (Nugroho et al. 2020).                                          VALD (Vienna Atomic Line Database, Piskunov et al. 1995).
     Driven by the intriguing characteristics of the system, in       We used a stellar mask with Teff =7500 K, logg=4.0, and so-
this work we analyse new transits of WASP-33b taken with the          lar metallicity. We accurately re-normalised the spectra (order-
HARPS-N high-resolution spectrograph, looking for confirma-           by-order by using polynomials, see Rainer et al. (2016) for
tion of the nodal precession and exploring its atmosphere. This       a detailed description of the procedure) and converted them
manuscript is organized as follows. We first present our data in      to the required format, working only on the wavelength re-
Sect. 2, then analyse the planetary Doppler shadow and the in-        gions 441.5 − 480.5 nm, 491.5 − 528.5 nm, 536.5 − 587.0 nm,
transit RVs studying also the precession of the orbital plane and     605.0 − 626.5 nm, and 633.5 − 645.0 nm, i.e., cutting the blue
of the spin of the host star (Sect. 3 and 4). We study the plane-     orders where the signal-to-noise ratio (S/N) was very low due to
tary atmosphere focusing on Hα and Hβ absorption in Sect. 5,          the instrument efficiency, the Balmer lines, and the regions where
finding a pre-transit signal coherent with the planetary orbital      most of the telluric lines are found. We then created mean line
period discussed in Sect. 6. Then we look for other atmospheric       profile residuals by dividing all the mean line profiles by a mas-
species with the cross-correlation technique in Sect. 7, ending       ter out-of-transit mean line profile for each transit observed. As
with summary and conclusions in Sect. 8.                              already evidenced by Collier Cameron et al. (2010) and Johnson
                                                                      et al. (2015), the Doppler shadow of the planet is clearly visible
                                                                      as well as the stellar pulsations (Fig. 1, left panel). We note that
2. Data sample                                                        the pulsations are more evident on the edge of the lines with re-
                                                                      spect to the center, which is a hint of non-radial pulsations (see
We observed WASP-33 during four transits using the high-              also Collier Cameron et al. 2010).
resolution (resolving power R∼115000) HARPS-N spectrograph                 For planets with projected spin-orbit angles close to 90 deg,
(Cosentino et al. 2012), mounted at the Telescopio Nazionale          nodal precession can be more easily detected during observa-
Galileo on the La Palma island and covering the wavelength            tions spanning several years (Watanabe et al. 2020). WASP-
range 380-690 nm. The first two transits were observed with           33b is one of the two reported cases where the exoplanet orbit
HARPS-N only (program A34TAC42, PI Nascimbeni), while                 has been found to show nodal precession (Johnson et al. 2015;
the other two were taken in the framework of the GAPS project         Watanabe et al. 2020), with the other case being Kepler-13Ab
(Covino et al. 2013). The latter were observed using the GIA-         (Szabó et al. 2012; Herman et al. 2018). Since our dataset ex-
RPS configuration (Claudi et al. 2017), but in this manuscript        tends the timespan of the reported orbital variations, we tried to
we focus only on the HARPS-N data.                                    see if this was visible also in our data. We removed stellar pulsa-
     The transit of WASP-33b lasts 2 hours 48 minutes. Three          tions independently for each single transit by using the Fourier
transit observations were monitored with exposures of 600 sec,        transform filtering (Fig. 1), following the method presented in
while for one we used a longer cadence (900 sec exposures).           Johnson et al. (2015). This exploits the fact that pulsations (pro-
While the fiber A of the spectrograph was centered on the target,     grade) and Doppler shadow (retrograde) propagate in opposite
for all the transits the fiber B was monitoring simultaneously the    directions: their frequency components thus tend to be sepa-
sky to check for possible atmospheric emission contamination.         rated in the two-dimensional Fourier transform of the line profile
The log of the observations is reported in Table 1.                   residuals time series (Johnson et al. 2015). We then performed a
     The quality of transit 2 rapidly decreases after the transit     fit to the Doppler shadow. The Doppler shadow model is taken
ingress due to weather conditions, we thus decided to not include     from EXOFASTv2 (Eastman 2017; Eastman et al. 2019), passed
it in the analysis of the transmission spectrum, Doppler shadow       through the Fourier filter and fitted to the data in a Bayesian
Article number, page 2 of 18
F. Borsa et al.: The WASP-33 system revisited with HARPS-N

Fig. 1. Tomography of the pulsation filtering from the mean line profile residuals. From top to bottom: transits 1, 3 and 4, respectively. (Left
panels): contour plot of the mean line profile residuals. (Central panels): same as for left panels, but after filtering in the Fourier space. (Right
panels): same as for central panels, after subtracting the best fit Doppler shadow model. The horizontal white lines show the beginning and end of
the transit.

framework by employing a differential evolution Markov chain                we report the systematic deviations in the derived model param-
Monte Carlo (DE-MCMC) technique (Ter Braak 2006; Eastman                    eters corresponding to a shift of the epoch of dλ/dt = 0 by ± 500
et al. 2013), running five DE-MCMC chains of 100,000 steps                  days since the assumed epoch, respectively (Table A.1). To com-
and discarding the burn-in. We fixed v sin i s , a/Rs , T0 , period,        pute these deviations, we assume that di/dt has the same value
Rp /Rs to the values of Table 2. We left as free parameters the             as above, while we linearly interpolate for the parameter i and
projected spin-orbit angle λ and the inclination angle i, for which         cubicly interpolate for the parameter λ to find their values at the
we set uninformative priors. The medians and the 15.86% and                 two epochs. We report for each parameter the standard deviation
84.14% quantiles of the posterior distributions were taken as the           as a measure of the statistical error, while we add in brackets the
best values and 1σ uncertainties.                                           systematic deviation when dλ/dt = 0 at the later epoch and the
    Our results (Fig. 2, Table 2), together with the previous mea-          deviation when dλ/dt = 0 at the earlier epoch, respectively. Note
surements of Watanabe et al. (2020), confirm the precession of              that the two deviations have the same sign when the parameter
the planetary orbit and that the obliquity projected on the plane           reached an extremum in between the considered epoches.
of the sky λ reached a minimum in 2011. This allows us to deter-                Following the method described in Appendix A, we find that
mine the inclination of the stellar spin axis to the line of sight is       the stellar spin is perpendicular to the line of sight (is = 90.11◦ ±
and the obliquity  at the epoch when dλ/dt = 0, thus allowing a            0.12◦ (−0.026◦ ; −0.015◦ )), while the obliquity  = 113.99◦ ±
more precise description of the precession of the system and the            0.22◦ (−0.24◦ ; −0.67◦ ) at the epoch of the transit on 19 October
measure of the stellar gravitational quadrupole moment J2 . We              2011. From the spectroscopic stellar v sin is , is , and stellar radius,
describe the method applied to determine the system geometry                we estimate a rotation period of the star of 0.884 ± 0.02 days, not
as well as the rate of precession of the nodes of the orbital plane         significantly affected by the systematic errors on is because it is
in Appendix A.                                                              close to 90◦ . The precession rate of the nodes of the orbital plane
    The rate of change of the inclination angle of the orbital              is found to be −0.325 ± 0.006 (6.7 × 10−5 ; 1.1 × 10−4 ) deg/yr
plane i can be regarded as constant over the time span of the               giving a precession period of 1108 ± 19 (0.23; 0.38) yr, slightly
observations (∼ 10 yr) because it is much shorter than the pre-             longer than that determined by Johnson et al. (2015). The time
cession period ( ∼ 1100 yr; see below). It is measured by a                 interval during which transits by WASP-33b are observable is
weighted linear best fit to the data in Fig. 2 (upper panel) giving         thus of ∼ 97 yr centred around 2019, slightly longer than their
di/dt = 0.324 ± 0.006 deg/yr, while the epoch when dλ/dt = 0                interval of about ∼ 88 yr.
is assumed to coincide with the transit observed on 19 Octo-                    Our inclination is of the stellar spin axis to the line of sight
ber 2011, when i = 87.56◦ ± 0.037◦ and λ = −114.01◦ ± 0.22◦                 is different from the value found by Iorio (2016) because his
(Watanabe et al. 2020). Given the uncertainty on such an epoch,             determination was based on the parameters derived only from the
                                                                                                                       Article number, page 3 of 18
A&A proofs: manuscript no. w33_acc

transits of 2008 and 2014 as reported by Johnson et al. (2015).
This led Iorio to consider a constant precession rate of the angle
I of his model, while our more extended dataset shows that dI/dt
is variable and became zero around 2011. In Appendix A.5, we
account for the differences between his results and ours and show
how the application of his model to our dataset reproduces our
value of is and of the stellar quadrupole moment (see below).
     Previous analyses of the precession of WASP-33b by John-
son et al. (2015) and Watanabe et al. (2020) assumed that the
stellar spin angular momentum is much larger than the orbital
angular momentum because they adopted the gyration radius γ
typical of a Sun-like star following the exploratory calculations
by Iorio (2011). Here we determine more appropriate values of
the gyration radius and apsidal motion constant k2 of WASP-33
by making use of the tabulations of Claret (2019). Considering
the uncertainties in the stellar mass and effective temperature,
we find that the Claret’s model with solar metallicity and mass
of 1.6 M is adequate to describe WASP-33. Taking into account            Fig. 2. Measurements of the inclination angle i (top panel) and of the
the correction for its fast rotation, we find log k2 = −2.55 ± 0.014     projected spin-orbit angle λ (bottom panel) as a function of time. The
and γ = 0.1884 ± 0.0019, where the uncertainties take into ac-           dashed line shows the linear trend discussed in Sect. 3.
count only the uncertainty in the stellar effective temperature.
     With the above values of k2 , γ, and the adopted stellar and        Table 2. Stellar and orbital parameters adopted in this work.
planetary parameters, we predict a value of di/dt = 0.315 ±
0.026 (0.0023; 0.0068) deg/yr and a precession rate of the nodes
                                                                                  Parameter              Value              Reference
of −0.316 ± 0.026 (0.0024; 0.0069) deg/yr using the formulae
given in Appendix A.3. The coincidence of these numerical val-                                         WASP-33
ues comes from the geometry of the system as explained there.
                                                                                                 Stellar parameters
These precession rates differ by less than a half standard devi-
ation from the observed ones, supporting the correctness of the                     T eff [K]       7430 ± 100            Yan et al. 2019
adopted stellar parameters. With our value of γ and the adopted                       log g            4.3 ± 0.2          Yan et al. 2019
                                                                                    [Fe/H]            -0.1 ± 0.2          Yan et al. 2019
system parameters, we find that the ratio of the stellar spin to                    Rs [R ]        1.509 ± 0.025          Yan et al. 2019
the orbital angular momentum is 2.75 ± 0.33, therefore, we can-                    Ms [M ]          1.561 ± 0.06          Yan et al. 2019
not neglect the precession of the stellar spin. It occurs with the             v sin i s [ km s−1 ]  86.4 ± 0.5             This work
same period as the precession of the node of the orbital plane and                Prot [days]       0.884 ± 0.02            This work
makes the inclination is of the stellar spin axis to the line of sight                           Orbital parameters
vary between 67.5◦ and 110.8◦ along a complete precession cy-
                                                                                 Period [days]     1.219870897           Yan et al. 2019
cle. Note that an inclination of the stellar spin greater than 90◦
                                                                              T 0 [BJD-2450000]     4163.22449           Yan et al. 2019
implies that the South pole of the star is in view, so the apparent                   Rp /Rs           0.11177           Yan et al. 2019
stellar rotation is clockwise, contrary to the usual anticlockwise                     a/Rs          3.69 ± 0.05         Yan et al. 2019
rotation assumed when is < 90◦ and the North pole is in view.                           e                 0.0               assumed
     The fast stellar rotation produces a deformation in the stel-                 i [degrees]      88.97 ± 0.04       This work, transit 1
lar mass distribution and hence a distortion in the gravitational                  i [degrees]      89.40 ± 0.05       This work, transit 3
field. The stellar gravitational quadrupole moment J2 can be de-                   i [degrees]      89.89 ± 0.03       This work, transit 4
duced from the observed nodal precession rate of the orbital                       λ [degrees]    −111.59 ± 0.25       This work, transit 1
plane according to eq. (3) of Johnson et al. (2015) and is (6.73 ±                 λ [degrees]    −111.70 ± 0.39       This work, transit 3
0.22 [0.062; 0.18])×10−5 . This value compares well with the the-                  λ [degrees]    −110.83 ± 0.30       This work, transit 4
                                                                                 Vsys [ km s−1 ]   −2.76 ± 0.04            This work
oretically expected value of (6.53 ± 0.41 [0.011; 0.031]) × 10−5
                                                                                  Kp [ km s−1 ]         231 ± 3          Yan et al. 2019
(e.g., Ragozzine & Wolf 2009, and Appendix A.4), thus confirm-                     Mp [MJup ]        2.16 ± 0.20         Yan et al. 2019
ing the goodness of the adopted values of k2 and γ. It also sup-                     Teq [K]         2710 ± 50           Yan et al. 2019
ports the hypothesis that WASP-33b is the only close massive
planet in the system because another similar body would con-
tribute to the orbital precession rate, if its orbit were not copla-
nar with that of WASP-33b. The fast rotation of WASP-33 makes
its quadrupole moment ∼ 375 times larger than the solar value            profile (Gray 2008). We measured a v sin i s =86.4 ± 0.5 km s−1 ,
(J2 ∼ 1.8 × 10−7 ; Rozelot et al. 2009), the effect of the cen-          taken as the average of the out-of-transit measurements and their
trifugal force being only modestly compensated by the stronger           standard deviation, which is well in agreement with previous val-
density concentration in an A-type star.                                 ues (v sin i s =86.63 km s−1 , Johnson et al. 2015). The extracted
                                                                         RVs are listed in Table B.1. We subtracted from each transit ob-
4. Radial velocities                                                     servation the difference from the mean of the in-transit RVs, to
                                                                         avoid any possible offset caused by instabilities, long-term pul-
Since we could not exploit the Cross-Correlation Functions               sations and trends in the orbital solution (e.g., Borsa et al. 2019),
(CCFs) from the DRS, RVs were extracted following the same               and then averaged the RVs of the four transits in bins of 0.005
method presented in Borsa et al. (2019). Instead of using a Gaus-        in phase. The RV time series of each single transit analysed and
sian fit, we preferred to model the LSD lines with a rotational          their average are shown in Fig. 3.
Article number, page 4 of 18
F. Borsa et al.: The WASP-33 system revisited with HARPS-N

               -1.5                                                                   -1.5
                                                              transit 1
                                                              transit 3
                                                              transit 4
               -2.0                                                                   -2.0

               -2.5                                                                   -2.5
   rv [km/s]

                                                                          rv [km/s]
               -3.0                                                                   -3.0

               -3.5                                                                   -3.5

               -4.0                                                                   -4.0
                      -0.10   -0.05   0.00       0.05         0.10                           -0.10   -0.05      0.00         0.05        0.10
                                       phase                                                                     phase

Fig. 3. (Left panel) Phase folded RVs of the three analysed HARPS-N transits of WASP-33b. Different colours refer to different transits. (Right
panel) RVs of the three transits averaged in bins of 0.005 in phase (filled circles), together with single RVs (small dots). The blue line represents
the theoretical RV solution calculated with the parameters of Table 2 (the values of λ and i are obtained from the average of the three transits).

     Although stellar pulsations dominate also the RV curve, it is               the airmass corresponding to that at the center of the transit. We
evident a qualitative agreement with the theoretical RM curve                    then created a master stellar spectrum S master by averaging all the
calculated using the parameters presented in Table 2 (by using                   out-of-transit spectra (excluding the phase range of the immedi-
the formalism of Ohta et al. 2005). This is a further confirma-                  ate pre-transit, see Sect. 6), and divided all the single spectra for
tion that for fast rotators the method of fitting the mean line                  this S master creating the residual spectra S res . Every S res was then
profiles with a rotational profile instead of a Gaussian function                shifted considering the theoretical planetary RV and the systemic
brings optimal results, as it was previously found for other tar-                velocity of the system, i.e., we placed the spectra in the plane-
gets (e.g., Anderson et al. 2018; Johnson et al. 2018; Borsa et                  tary reference frame. Here we expect to detect the exoplanetary
al. 2019; Rainer et al. 2021). We note however that for this case                atmospheric signal centered at the laboratory wavelengths. All
the Doppler tomography method remains the preferred one to                       the full-in-transit S res were then averaged to create the transmis-
determine the projected spin-orbit inclination angle.                            sion spectrum. At this stage, the transmission spectrum of the
                                                                                 planet still includes spurious stellar contaminations.

5. Balmer lines absorption in the transmission
   spectrum                                                                      5.2. Removal of CLV + RM

We studied the atmosphere of WASP-33b by using the technique                     The star over which the planet transits is not a simple homo-
of transmission spectroscopy, focusing on Hα and Hβ absorption                   geneous disk, but rotates and has a surface brightness which
and taking into account the possible contamination given by stel-                changes as a function of the distance from center. Effects such
lar effects.                                                                     as center-to-limb variations (CLV) and RM have been proven
                                                                                 to significantly modify the shape of line profiles, possibly caus-
5.1. Transmission spectrum extraction                                            ing false atmospheric detections (e.g., Yan et al. 2017; Borsa &
                                                                                 Zannoni 2018; Casasayas-Barris et al. 2020). We thus took these
We performed transmission spectroscopy following the method                      effects into account by creating a model following the method-
of Wyttenbach et al. (2015), applying the following steps in-                    ology described in Yan et al. (2017). The star is modeled as a
dependently to each transit. First we shifted the spectra to the                 disk divided in sections of 0.01 Rs . For each point, we calcu-
stellar radial velocity rest frame using a Keplerian model of the                late the µ value (where µ = cos θ, with θ the angle between
system. Then we normalized each spectrum, by dividing for the                    the normal to the stellar surface and the line of sight) and the
average flux within defined wavelength ranges where telluric and                 projected rotational velocity (by assuming rigid body rotation
stellar lines are not present.                                                   and rescaling the v sin i s value of Table 2). A spectrum is then
    Using the out-of-transit spectra only, we built a telluric refer-            assigned to each point of the grid, by quadratically interpolat-
ence spectrum T (λ), by means of a linear correlation between the                ing on µ and Doppler-shifting for the stellar rotation the model
logarithm of the normalized flux and the airmass (Snellen et al.                 spectra created using the tool Spectroscopy Made Easy (SME,
2008; Vidal-Madjar et al. 2010; Astudillo-Defru & Rojo 2013),                    Piskunov & Valenti 2017), with the line list from the VALD
and then rescaled all the spectra as if they had been observed at                database (Ryabchikova et al. 2015) and the MARCS (Gustafs-
                                                                                                                           Article number, page 5 of 18
A&A proofs: manuscript no. w33_acc

                                                                           lines largely affects the wings, but not so much the center, where
                                                          CLV+RM MODEL
                  1.002                                                    we search for the in-transit planetary RVs: Balmer and metallic
                                                          PULSATIONS
                                                                           line-profiles are very similar there. We also note that we can-
                                                                           not analyse the pulsation pattern directly on the Balmer lines,
                  1.001                                                    due to their low S/N. As a final test we verified that the applied
                                                                           correction does not impact the depth of the retrieved planetary
Normalized flux

                                                                           line-profile (Sect. 5.4).
                  1.000
                                                                               We started from the mean line profiles residuals, after the
                                                                           division for an average out-of-transit mean line profile. We then
                  0.999
                                                                           subtracted for each transit the Doppler shadow models calculated
                                                                           in Sect. 3 (e.g., Fig. 5). Then we created a transmission spectrum
                                                                           of the pulsations (TS puls ) in the wavelength region of interest.
                  0.998                                                    First we put together the three transits, then we move from the
                                                                           velocity to the wavelength space, taking as zero reference the
                                                                           laboratory wavelength of the Hα (Fig. 6) and Hβ lines, since
                          6560   6562          6564        6566            these are the lines we aim at correcting. Then we moved each
                                 Wavelength [Angstrom]
                                                                           mean line profile residual to the planetary restframe, and aver-
                                                                           aged the full-in-transit residuals to calculate the TS puls . At this
Fig. 4. The corrections for CLV+RM and for stellar pulsations applied
to the transmission spectrum in the Hα zone.
                                                                           point we still have to correct for the magnitude of the effect: the
                                                                           calculated TS puls is in fact referred to a line whose depth is the
                                                                           one of the mean line profile, and has to be rescaled to the depth
                                                                           of the line that we want to correct (Borsa & Zannoni 2018). We
son et al. 2008) stellar atmospheric models (assuming local ther-          thus multiplied the amplitude of the correction for a factor which
modynamic equilibrium approximation). The model spectra are                is the ratio between the depth of the mean line profile and that
created with null rotational velocity for 21 different µ values,           of the Hα (or Hβ) line, and obtained the final spectrum used to
and adapted to the resolving power of the instrument. Then, us-            correct for pulsations (Fig. 4).
ing the orbital information from Table 2 we simulate the transit
of the planet, calculating the stellar spectrum for different orbital
phases as the average spectrum of the non-occulted modeled sec-            5.4. Balmer lines absorption
tions. As the last step, we divide each spectrum for a master stel-
lar spectrum calculated out-of-transit, obtaining the information          We corrected the transmission spectrum extracted in Sect. 5.1
of the impact of RM+CLV effects at each in-transit orbital phase.          from stellar effects by dividing it for the corrective spectra cal-
We then move everything in the planetary restframe and calcu-              culated in Sect. 5.2 and 5.3. We note that the effect of pulsations
late the simulated CLV+RM effects on the transmission spec-                overcomes that of CLV+RM by a factor ∼1.5 in magnitude (Fig.
trum (Fig. 4). We note that we do not take into account the pos-           4). We find an excess absorption in the Hα region with a contrast
sible deformation of the line profiles given by gravity darken-            of 0.54 ± 0.04% and a FWHM of 36.7 ± 3.3 km s−1 (Fig. 7).
ing, which are however not expected to be significant (see, e.g.,          We can translate the contrast into an effective planetary radius
Kovács et al. 2013).                                                       Reff =1.18 ± 0.02 Rp , calculated assuming R2eff /R2p = (δ + h)/δ,
                                                                           with δ the transit depth (from Table 2) and h the line contrast
                                                                           (e.g., Chen et al. 2020). As for KELT-9b, the Hα profile of
5.3. Removal of pulsations                                                 WASP-33b forms in the atmospheric region below the Roche
                                                                           lobe, which lies at about 1.6 planetary radii. Therefore this detec-
Another contamination one has to deal with in this particular              tion does not directly probe atmospheric escape, but it would en-
case of WASP-33b comes from the stellar pulsations. We already             able one to set constraints on the planetary temperature-pressure
showed in Sect. 3 how strongly they impact the mean line pro-              structure and thus on the energetics possibly driving escape. The
files. In the same way, they will have an impact in the extracted          line has a significant blueshift of -8.2 ± 1.4 km s−1 , indicative
planetary transmission spectrum. To deal with this, we used an             of winds in the planetary atmosphere. We note that without the
approach similar to that presented in Borsa & Zannoni (2018)               correction for the stellar pulsations we find a comparable depth
to remove the RM from the transmission spectrum by using real              (0.53%) but a larger FWHM (55 km s−1 ). Being non-radial pul-
observing data. This was done by exploiting the S/N of the mean            sations, their averaged effect on the three transits tends to be
line profiles. We note that pulsations and RM cannot be removed            smaller in the center of the stellar line profile, and this is re-
together in this way. This is because the amplitude of the RM ef-          flected also in the transmission spectrum. This is also noticeable
fect is dependent on the planetary effective radius in the narrow          in Fig. 4 and Fig. 6.
bandpasses where the planetary lines are present. On the con-                   Checking the reference frame of the detections is fundamen-
trary, the amplitude of pulsations vary on a much larger wave-             tal when we want to be sure they are caused by the planetary
length range and can be assumed constant in each of our narrow             atmosphere and not by spurious stellar effects (e.g., Brogi et al.
bandpasses of interest. In the following, we make the assump-              2016; Borsa & Zannoni 2018), in particular for this case where
tion that the pulsation pattern behaves in the same way for all            the strong stellar pulsations could be not perfectly corrected for
the absorption lines. We use the pulsation pattern extracted from          and mimic atmospheric features. We thus verified that our Hα
the mean line profile (which is basically based on metal lines)            absorption detection is in the planetary restframe by resolving it
and assume that it is valid also for the Balmer lines, except for a        in the 2D tomographic map (Fig. 8), which shows the position
scaling factor. To justify this assumption we note that the short          of the absorption signal as a function of the orbital phase.
periods of δ Sct are not able to produce significant phase shifts               We also find an excess absorption in the Hβ line region
between the photospheric stellar lines of different species and            with a contrast of 0.28±0.06%, FWHM of 23.5±6.3 km s−1
excitation potentials. Moreover, the broadening of the Balmer              and a blueshift of -6.6±2.6 km s−1 (Fig. 9 and Fig. 10). We
Article number, page 6 of 18
F. Borsa et al.: The WASP-33 system revisited with HARPS-N

Fig. 5. Example of the removal of the Doppler shadow from the mean line profile residuals for transit 4. (Left panel) Original mean line profile
residuals. (Central panel) Model of the Doppler shadow. (Right panel) Mean line profile residuals after the removal of the Doppler shadow. The
horizontal white lines show the beginning and end of the transit.

                                                                                                     1.010

                                                                                                     1.005
                                                                              planetary absorption

                                                                                                     1.000

                                                                                                     0.995

                                                                                                     0.990

                                                                                                     0.985

                                                                                                         6555   6560                    6565          6570
                                                                                                                       Wavelength [Å]

                                                                             Fig. 7. Hα absorption after the correction for stellar effects (RM+CLV
                                                                             and pulsations). Black circles represent 0.1 Å binning. The blue line is
                                                                             the Gaussian best fit. The vertical dotted line shows the planetary Hα
Fig. 6. The effect of the pulsations contamination on the three analysed     restframe.
transits as it affects the Hα line, in the stellar restframe. The vertical
dotted line marks the position of the Hα line. The horizontal white lines
show the beginning and end of the transit.                                   activity (e.g., Borsa et al. 2021) and/or pulsations. We checked
                                                                             within the dataset we analysed, without finding evidence of sig-
                                                                             nificant variability with all the three transits having the same Hα
can translate the contrast into an effective planetary radius                contrast.
Reff =1.09±0.03 Rp .                                                             A future common analysis of all available datasets will help
     We note that the values of absorption depth we obtain are               shedding light on the possibility of the observed transit depth
almost half of those obtained for the same planet by Yan et                  variations being related to stellar activity and/or pulsation, rather
al. (2021) (0.99±0.05% for Hα and 0.54±0.07% for Hβ, re-                     than caused by systematics between different analyses. Further-
spectively). However, the Hα/Hβ line depth ratio is the same                 more, such an analysis will enable one to obtain a very high-
(1.93±0.44 for us versus 1.83±0.26 for them). Given the fact                 quality average line profile for each detected Balmer line. These
that also another analysis of these lines shows different line pro-          profiles can then be used to constrain the temperature structure of
file depths (1.68±0.02% for Hα and 1.02±0.05% for Hβ, Cauley                 the planetary atmosphere in the 10−3 –10−9 bar range (e.g., Fos-
et al. 2021), we highlight the possibility that the amplitude of the         sati et al. 2020), going beyond the analysis of Yan et al. (2021),
lines could be variable with time and with the level of the stellar          who employed an isothermal profile and assumed local thermo-
                                                                                                                                 Article number, page 7 of 18
A&A proofs: manuscript no. w33_acc

                                                                             a thermal inversion in the upper atmosphere and reach upper at-
                                                                             mospheric temperatures of the order of 10000 K; Fossati et al.
                                                                             2018, 2020), it is likely that the assumptions employed by Yan
                                                                             et al. (2021) for modelling the Balmer lines led them to obtain
                                                                             unreliable results.

                                                                             6. Pre-transit signal
                                                                             Looking at the tomographic map of the Hα absorption (Fig. 8),
                                                                             we noted a pre-transit feature. This feature has a different slope
                                                                             and duration in the 2D tomographic map with respect to the dom-
                                                                             inant pulsation pattern in the mean line profiles (see Fig. 1 and
                                                                             Fig. 11). We checked one by one the single transits, and found
                                                                             that this feature looks quite similar in all four available transits
Fig. 8. Contour 2D tomography map of Hα absorption in the stellar (left      (Fig. 11, see the blue diagonal track before the beginning of the
panel) and planetary (right panel) restframes, before applying the stellar   transit). It is thus something which happens in an almost coher-
contamination correction (this evidences also the Doppler shadow, i.e.,
                                                                             ent timing with the orbital period of the planet.
the red track). The white horizontal lines represent the beginning and
end of the transit. The vertical dotted black line shows the Hα planetary         It appears to be a pseudo-absorption pre-transit signal (PTS
restframe in the right panel and the stellar restframe in the left panel.    hereafter) happening on the stellar surface, because it is moving
A pre-transit signal, not centered in the planetary restframe, is evident    from -v sin i s to +v sin i s (i.e., across the stellar CCF) with time.
(see discussion in Sect. 6).                                                 We thus investigated this more in detail, performing the normal-
                                                                             ization on an extended out-of-transit baseline to avoid including
                                                                             the PTS in it. By visual inspection, we noted that a similar be-
                                                                             haviour is observed in many single lines, not only on Hα. On the
                                                                             mean line profiles it is hidden by the other stellar pulsations, but
                                                                             is still noticeable after removing them (Fig. 1, right panel). Cu-
                                                                             riously, this feature is also reflected in the radial velocities (see
                                                                             Fig. 3).
                                                                                  We found that the restframe of the PTS is not compatible
                                                                             with the one of the planet, by studying it on the Hα line and as-
                                                                             suming it is moving with a Keplerian motion around the star with
                                                                             the same orbital period of the planet. Thus we define KPTS as the
                                                                             Keplerian semi-amplitude of this pseudo-orbital motion. By fit-
                                                                             ting Gaussian functions and maximizing the contrast of their av-
                                                                             erage, we could estimate that the PTS is moving at a KPTS ∼
                                                                             460 km s−1 and is centered at phase ∼-0.07. As averaged on
                                                                             these values, the PTS has a contrast of 0.61±0.06 % (Fig. 12).
                                                                             Curiously, the value of the KPTS is almost double as the one of
                                                                             Kp (231 km s−1 , Table 2). The two shoulders around the profile
                                                                             (Fig. 12) and the red tracks around the signal (Fig. 11) mimic
Fig. 9. Same of Fig. 7 but for Hβ.                                           the presence of a Doppler shadow, but this could be possibly due
                                                                             to a normalisation artifact (e.g., Snellen et al. 2010). Yan et al.
                                                                             (2021) do not find any signs of pre-transit absorption in their Hα
                                                                             analysis of WASP-33b, but we note that they looked in the plan-
                                                                             etary restframe only, while we noticed this PTS while looking in
                                                                             the stellar restframe.
                                                                                  Many photometric transit observations of WASP-33b have
                                                                             been gathered (see Sect. 1), and even if they are affected by
                                                                             the stellar pulsations, none of them reported a pre-transit re-
                                                                             duction of flux coherent with the planetary orbital period. The
                                                                             most probable option in our opinion is that the PTS is still a stel-
                                                                             lar pulsation mode, which is possibly/probably excited by the
                                                                             planetary companion just before the transit. The PTS likely be-
                                                                             longs to the same pulsations pattern well visible on Hα and Hβ
                                                                             lines in Cauley et al. (2021, see their Fig. 5), and also notice-
                                                                             able in Fig. 1, right panel. When averaging different transits the
Fig. 10. Same of Fig. 8 but for Hβ.                                          PTS does not average out like the other pulsations, but tends to
                                                                             clearly stand out beyond the noise. This would be indeed ev-
                                                                             idence of a somewhat expected star-planet interaction for this
dynamical equilibrium. Indeed, it has been shown that, for mod-              system. Mechanisms that can excite stellar pulsations could have
elling the Balmer lines detected in the transmission spectrum of             tidal or magnetic origin.
the UHJ KELT-9b, these assumptions are invalid (Fossati et al.                    Planet-induced stellar pulsations were reported for exam-
2020). Because of similarities in the temperature profile of the             ple in the HAT-P-2 system, where de Wit et al. (2017) discov-
upper atmosphere of KELT-9b and WASP-33b (e.g., both present                 ered pulsation modes corresponding to exact harmonics of the
Article number, page 8 of 18
F. Borsa et al.: The WASP-33 system revisited with HARPS-N

Fig. 11. Zoom on the pre-transit signal on the 4 transits of our dataset, in the stellar restframe. Vertical line shows the Hα line center, while vertical
dotted lines mark the ±v sin i s limits. Horizontal white line shows the beginning of the transit. The pre-transit signal is the diagonal blue track
before the beginning of the transit.

                                                                               ics of the orbital frequency, were proposed also for the systems
                  1.005
                                                                               WASP-12 and WASP-18 (Maciejewski et al. 2020a,b). Tidally
                                                                               induced flux modulations are also shown in heartbeat stars (e.g.,
                                                                               Welsh et al. 2011; Thompson et al. 2012), caused by large hy-
                                                                               drostatic adjustments due to the strong gravitational distortion
                                                                               they experience during the periastron passage of the eccentric
Normalized flux

                  1.000
                                                                               binary companion. We thus can hypothesize a tidal star-planet
                                                                               interaction to be the cause of the PTS. When considering equi-
                                                                               librium tides, we should find two RV maxima/minima during
                                                                               the planetary orbit, but the effect should be of the order of ∼10
                  0.995                                                         m s−1 (i.e., lower than the one seen for WASP-18 considering
                                                                               the masses and orbital configuration of the two systems), thus
                                                                               not detectable. When dealing with pulsations excited by tides
                                                                               instead, their amplitude could be much larger, but the lack of a
                          -400   -200        0           200          400      stellar pulsation frequency close to the expected period of tides
                                        vel [km/s]
                                                                               (∼1.6 days, assuming a stellar rotational period of 0.884 days
Fig. 12. Pre-transit signal profile in its restframe with KPTS =460 km s−1 .   and a planetary orbital period of 1.22 days) in the frequencies
The blue line shows the best-fit Gaussian profile.                             shown in von Essen et al. (2020) is not in favour of this hy-
                                                                               pothesis. Moreover, the duration of a resonance between pulsa-
                                                                               tions and tides should be very short with respect to the evolutive
                                                                               timescales of the system. The large oscillations produced by the
planet’s orbital frequency, indicative of a tidal origin. Planetary            resonance would in fact dissipate a lot of energy, producing an
induced tides in the host star, manifesting as the second harmon-
                                                                                                                           Article number, page 9 of 18
A&A proofs: manuscript no. w33_acc

exchange of angular momentum between rotational and orbital             e2ds HARPS-N spectra order-by-order. We define the cross-
motion that modifies both, moving the system out of the reso-           correlation as
nance soon. So the probability that we are observing the system
exactly in this moment is low, which is also not favouring the                      N
                                                                                    X
tidal hypothesis.                                                       C(v, t) =         xi (t, v)T i                                   (1)
    Another possibility for the detected PTS is that of a mag-                      i=1
netic interaction. Magnetic star-planet interaction has been pro-       where T is the model template normalized to unity and x are the
posed and observed for some short-period systems, mainly as             normalized fluxes at the N wavelengths i of the spectra taken at
the modulation of stellar activity index with the orbital period        time t and shifted at velocity v. In this way we preserve the flux
(e.g., Lanza 2009, 2012; Strugarek 2018; Cauley et al. 2019).           information (e.g., Hoeijmakers et al. 2019). In our models we
A pre-transit absorption in the Balmer lines was observed for           impose all the values with contrast less than 5% of the maximum
HD 189733b (Cauley et al. 2015) and was proposed as be-                 in the wavelength range to be zero.
ing caused by the planetary magnetic field, with a bow shock                We selected a step of 1 km s−1 and a velocity range [-
forming around the magnetosphere heading ahead of the planet            300,300] km s−1 , performing the cross-correlation for each or-
(Llama et al. 2013; Cauley et al. 2015). Magnetic fields are            der. We performed a 5σ-clipping and masked the wavelength
known to have effect on stellar pulsations, depending on their          ranges most affected by telluric contamination. Then for each
magnitude (e.g., Handler 2013, and references therein). The in-         exposure we applied a weighted average between the cross-
teresting possibility of the PTS being a magnetic excitation of         correlations of the single orders, where the weights applied to
pulsations given by the interaction with the planetary magnetic         each order are the inverse of the standard deviation (i.e., the
field could be further explored with spectro-polarimetric mea-          larger the S/N, the higher the weight) times the depths of the
surements, since magnetic fields affect not only the spectral line      lines in the model template. For a range of Kp values from 0 to
profiles but also polarization properties of stellar radiation (e.g.,
Landi Degl’Innocenti & Landolfi 2004; Oklopčić et al. 2020).          300 km s−1 (with steps of 1 km s−1 ) we averaged the in-transit
                                                                        cross-correlation functions after shifting them in the planetary
                                                                        restframe. This last step is done by subtracting the planetary ra-
                                                                        dial velocity calculated for each spectrum as v p = Kp × sin 2πφ,
                                                                        with φ the orbital phase. In this way, we created the Kp vs Vsys
7. Cross-correlation with templates                                     maps, that are used to test the planetary origin of any possible
                                                                        signal. We evaluated the noise by calculating the standard devi-
Since we found the KPTS to be almost double of the planetary            ation of the Kp vs Vsys maps far from where any stellar or plan-
KP (Sect. 6), we tried to test a possible relation of the PTS with      etary signal is expected, in particular where |V| > 90 km s−1 .
the planetary atmosphere, which could have left aliases or re-              We find no evidence supporting the presence in the planetary
flections/scattering in the spectra after removing the stellar mas-     atmosphere either of V i or AlO, giving upper limits (which are
ter. We thus looked in our planetary transmission spectrum for          model dependent) of 97 ppm and 17 ppm at the 3σ level, re-
species that could be present only in the planet atmosphere and         spectively. While for V i we are confident in the accuracy of the
not in the star, to be further investigated in the PTS restframe in     line list we used, as it has also already brought a clear detection
case of detection. We decided to look for Vanadium (V i), which         (Borsa et al. 2021), this is not the case for AlO. We performed in-
has been already detected in an exoplanetary atmosphere and             jection of our AlO model in our data using the abundance found
not in the host star (Ben-Yami et al. 2020; Borsa et al. 2021),         in the detection by von Essen et al. (2019), and we could re-
and for Aluminum oxide (AlO), which has been claimed to be              cover it with ∼14σ significance. Although we did not find evi-
present in the atmosphere of WASP-33b through low-resolution            dence for its presence with the available opacities, once an accu-
spectrophotometric observations (von Essen et al. 2019).                rate AlO linelist will be available verifying with high-resolution
     Planetary atmosphere transmission models were created by           spectroscopy the claimed presence of AlO in the planetary atmo-
using petitRADTRANS (pRT, Mollière et al. 2019). We assumed             sphere (von Essen et al. 2019) will be possible using this dataset.
solar abundances, a continuum pressure level of 1 mbar, and
an atmospheric temperature-pressure profile calculated for the
planet in the same way as presented in Fossati et al. (2020).           8. Summary and conclusions
These parameters were set because they are typical for UHJs             We provide further evidence that WASP-33 is undergoing nodal
(e.g., Hoeijmakers et al. 2019; Stangret et al. 2020). The at-          precession, finding a precession period of 1108 ± 19 yr, slightly
mospheric models, created in planetary radius as a function of          longer than that determined by Johnson et al. (2015). We also
wavelength, were translated into flux (Rp /Rs )2 , convolved at the     found that the stellar spin is perpendicular to the line of sight,
HARPS-N resolving power and continuum normalized. While                 and determined the gravitational quadrupole moment of the star
for V i we used the opacities already included within pRT, for          J2 = (6.73 ± 0.22) × 10−5 , which is in close agreement with
AlO we used those available at the opacity database of the Ex-          the theoretically predicted value of (6.53 ± 0.41) × 10−5 (e.g.,
oplanet Simulation Platform1 (Grimm & Heng 2015), that we               Ragozzine & Wolf 2009). By increasing the time coverage of
arranged in the pRT format.                                             WASP-33b spectroscopic transit observations it will be possi-
     Cross-correlation between the data and the models was per-         ble to constrain even more the precession period and possibly,
formed in the stellar restframe on single residual spectra after the    with the growing precision of new-generation instruments, also
removal of a master out-of-transit star and telluric contamination      to detect relativistic effects (Iorio 2016). It would be important to
(with the procedures explained in Sect. 5) on the whole wave-           look for other exoplanets in which we can detect orbital preces-
length range. This was done by working on the bi-dimensional            sion (other than WASP-33b and Kepler-13Ab), to increase the
                                                                        statistics and understand if this phenomenon could depend also
                                                                        on the presence of other companions, on the value of the stellar
1
    https://dev.opacity.iterativ.ch/                                    rotation or on the orbital obliquity of the system.
Article number, page 10 of 18
F. Borsa et al.: The WASP-33 system revisited with HARPS-N

     We found that stellar pulsations contaminate the extracted                     Gray, D.F., 2008, The Observation and Analysis of Stellar Photospheres, Cam-
transmission spectrum of the planet and proposed a method to                           bridge, UK: Cambridge University Press
detrend it, which can be potentially applied to mitigate also stel-                 Grimm, S. L. & Heng, K. 2015, ApJ, 808, 182
lar activity. The detected Hα absorption in the planetary atmo-                     Gustafsson, B., Edvardsson, B., Eriksson, K., et al. 2008, A&A, 486, 951
sphere extends up to ∼1.18 Rp , while absorption in Ca ii H&K                       Handler, G. 2013, Planets, Stars and Stellar Systems. Volume 4: Stellar Structure
was detected with an effective radius of ∼1.56 Rp (Yan et al.                          and Evolution, 207
2019). This confirms the tendency of Ca ii H&K and Hα to exist                      Haynes, K., Mandell, A. M., Madhusudhan, N., et al. 2015, ApJ, 806, 146
up to the highest layers of the atmospheres of hot Jupiters. The                    Herman, M. K., de Mooij, E. J. W., Huang, C. X., et al. 2018, AJ, 155, 13
line contrast of the planetary absorption we measured for both                      Herman, M. K., de Mooij, E. J. W., Jayawardhana, R., et al. 2020, AJ, 160, 93
Hα and Hβ lines is lower than previously found in the literature                    Herrero, E., Morales, J. C., Ribas, I., et al. 2011, A&A, 526, L10
(Yan et al. 2021; Cauley et al. 2021), while their contrast ratio                   Hoeijmakers, H. J., Ehrenreich, D., Kitzmann, D., et al. 2019, A&A, 627, A165
is the same. This opens the possibility that the atmosphere of                      Iorio, L. 2011, Ap&SS, 331, 485
WASP-33b could be sensitive to a variable level of activity of                      Iorio, L. 2016, MNRAS, 455, 207
the star.                                                                           Johnson, M. C., Cochran, W. D., Collier Cameron, A., et al. 2015, ApJ, 810, L23
     We detected a pre-transit signal almost coherent with the or-                  Johnson, M. C., Rodriguez, J. E., Zhou, G., et al. 2018, AJ, 155, 100
bital period of the planet. The most likely explanation is that                     Kovács, G., Kovács, T., Hartman, J. D., et al. 2013, A&A, 553, A44
this is a stellar pulsation mode, which is excited by the planetary                 Landi Degl’Innocenti, E., & Landolfi, M. 2004, Astrophysics and Space Science
                                                                                       Library, Vol. 307, Polarization in Spectral Lines
companion. The nature of this possible star-planet interaction is
                                                                                    Lanza, A. F. 2009, A&A, 505, 339
still doubtful, even if we tend to exclude tidal interactions and
                                                                                    Lanza, A. F. 2012, A&A, 544, A23
are more in favour of possible magnetic interactions. Spectro-
                                                                                    Leconte, J., Chabrier, G., Baraffe, I., et al. 2010, A&A, 516, A64
polarimetric observations of the transit and a complete spectro-
                                                                                    Lehmann, H., Guenther, E., Sebastian, D., et al. 2015, A&A, 578, L4
scopic phase coverage, as well as a detailed analysis of stellar
                                                                                    Llama, J., Vidotto, A. A., Jardine, M., et al. 2013, MNRAS, 436, 2179
pulsations on the line profiles, could help in understanding the
                                                                                    Lothringer, J. D., Barman, T., & Koskinen, T. 2018, ApJ, 866, 27
nature of this pre-transit signal.
                                                                                    Maciejewski, G., Niedzielski, A., Villaver, E., et al. 2020, ApJ, 889, 54
Acknowledgements. We thank the referee for their useful comments that helped        Maciejewski, G., Knutson, H. A., Howard, A. W., et al. 2020, Acta Astron., 70,
improving the clarity of the manuscript. We acknowledge the support by                 1
INAF/Frontiera through the "Progetti Premiali" funding scheme of the Italian
                                                                                    McLaughlin, D. B. 1924, ApJ, 60, 22
Ministry of Education, University, and Research and from PRIN INAF 2019.
FB acknowledges financial support from INAF through the ASI-INAF contract           Mkrtichian, D. 2015, IAU General Assembly 29, 2255391
2015-019-R0.                                                                        Mollière, P., Wardenier, J. P., van Boekel, R., et al. 2019, A&A, 627, A67
                                                                                    Nugroho, S. K., Kawahara, H., Masuda, K., et al. 2017, AJ, 154, 221
                                                                                    Nugroho, S. K., Gibson, N. P., de Mooij, E. J. W., et al. 2020, ApJ, 898, L31
                                                                                    Ogilvie, G. I. 2014, ARA&A, 52, 171
References                                                                          Ohta, Y., Taruya, A., & Suto, Y. 2005, ApJ, 622, 1118
Akinsanmi, B., Barros, S. C. C., Santos, N. C., et al. 2019, A&A, 621, A117         Oklopčić, A., Silva, M., Montero-Camacho, P., et al. 2020, ApJ, 890, 88
Anderson, D. R., Temple, L. Y., Nielsen, L. D., et al. 2018, arXiv e-prints,        Parmentier, V., Line, M. R., Bean, J. L., et al. 2018, A&A, 617, A110
   arXiv:1809.04897
                                                                                    Piskunov, N., Kupka, F., Ryabchikova, T. A., Weiss, W. W., & Jeffery, C. S.1995,
Arcangeli, J., Désert, J.-M., Line, M. R., et al. 2018, ApJ, 855, L30
                                                                                       A&AS, 112, 525
Astudillo-Defru, N., & Rojo, P. 2013, A&A, 557, A56
Ben-Yami, M., Madhusudhan, N., Cabot, S. H. C., et al. 2020, ApJ, 897, L5           Piskunov, N., & Valenti, J. A. 2017, A&A, 597, A16
Borsa, F., & Zannoni, A. 2018, A&A, 617, A134                                       Ragozzine, D. & Wolf, A. S. 2009, ApJ, 698, 1778
Borsa, F., Rainer, M., Bonomo, A. S., et al. 2019, A&A, 631, A34                    Rainer, M., Poretti, E., Mistò, A., et al. 2016, AJ, 152, 207
Borsa, F., Allart, R., Casasayas-Barris, N., et al. 2021, A&A, 645, A24             Rainer, M., Borsa, F., Pino, L., et al. 2021, A&A, 649, A29
Brogi, M., de Kok, R. J., Albrecht, S., et al. 2016, ApJ, 817, 106
                                                                                    Rossiter, R. A. 1924, ApJ, 60, 1
Casasayas-Barris, N., Palle, E., Yan, F., et al. 2020, A&A, 635, A206
Cauley, P. W., Redfield, S., Jensen, A. G., et al. 2015, ApJ, 810, 13               Rozelot, J. P., Damiani, C., & Pireaux, S. 2009, ApJ, 703, 1791
Cauley, P. W., Shkolnik, E. L., Llama, J., et al. 2019, Nature Astronomy, 3, 1128   Ryabchikova, T., Piskunov, N., Kurucz, R. L., et al. 2015, Phys. Scr, 90, 054005
Cauley, P. W., Wang, J., Shkolnik, E. L., et al. 2021, arXiv:2010.02118             Smith, A. M. S., Anderson, D. R., Skillen, I., et al. 2011, MNRAS, 416, 2096
Chen, G., Casasayas-Barris, N., Pallé, E., et al. 2020, A&A, 635, A171              Snellen, I. A. G., Albrecht, S., de Mooij, E. J. W., & Le Poole, R. S. 2008, A&A,
Claret, A. 2019, A&A, 628, A29                                                         487, 357
Claudi, R., Benatti, S., Carleo, I., et al. 2017, European Physical Journal Plus,
   132, 364                                                                         Snellen, I. A. G., de Kok, R. J., de Mooij, E. J. W., et al. 2010, Nature, 465, 1049
Collier Cameron, A., Bruce, V. A., Miller, G. R. M., et al. 2010, MNRAS, 403,       Stangret, M., Casasayas-Barris, N., Pallé, E., et al. 2020, A&A, 638, A26
   151                                                                              Strugarek, A. 2018, Handbook of Exoplanets, 25
Collier Cameron, A., Guenther, E., Smalley, B., et al. 2010, MNRAS, 407, 507        Szabó, G. M., Pál, A., Derekas, A., et al. 2012, MNRAS, 421, L122
Cosentino, R., Lovis, C., Pepe, F., et al. 2012, Proc. SPIE, 8446, 84461V           Ter Braak, C. J. F. 2006, Statistics and Computing, 16, 239
Cosentino, R., Lovis, C., Pepe, F., et al. 2014, Proc. SPIE, 9147, 91478C
Covino, E., Esposito, M., Barbieri, M., et al. 2013, A&A, 554, A28                  Thompson, S. E., Everett, M., Mullally, F., et al. 2012, ApJ, 753, 86
Damiani, C. & Lanza, A. F. 2011, A&A, 535, A116                                     Vidal-Madjar, A., Arnold, L., Ehrenreich, D., et al. 2010, A&A, 523, A57
de Wit, J., Lewis, N. K., Knutson, H. A., et al. 2017, ApJ, 836, L17                von Essen, C., Czesla, S., Wolter, U., et al. 2014, A&A, 561, A48
Donati, J.-F., Semel, M., Carter, B. D., Rees, D. E., & Collier Cameron, A. 1997,   von Essen, C., Mallonn, M., Albrecht, S., et al. 2015, A&A, 584, A75
   MNRAS, 291, 658
                                                                                    von Essen, C., Mallonn, M., Welbanks, L., et al. 2019, A&A, 622, A71
Eastman, J., Gaudi, B. S. & Agol, E. 2013, PASP, 125, 923
Eastman, J. 2017, Astrophysics Source Code Library, ascl:1710.003                   von Essen, C., Mallonn, M., Borre, C. C., et al. 2020, A&A, 639, A34
Eastman, J. D., Rodriguez, J. E., Agol, E., et al. 2019, arXiv:1907.09480           Watanabe, N., Narita, N., & Johnson, M. C. 2020, PASJ, 72, 19
Fabrycky, D. C. & Winn, J. N. 2009, ApJ, 696, 1230                                  Welsh, W. F., Orosz, J. A., Aerts, C., et al. 2011, ApJS, 197, 4
Fossati, L., Koskinen, T., Lothringer, J. D., et al. 2018, ApJ, 868, L30            Wyttenbach, A., Ehrenreich, D., Lovis, C., et al. 2015, A&A, 577, A62
Fossati, L., Koskinen, T., France, K., et al. 2018, AJ, 155, 113
Fossati, L., Shulyak, D., Sreejith, A. G., et al. 2020, A&A, 643, A131.             Yan, F., Pallé, E., Fosbury, R. A. E., et al. 2017, A&A, 603, A73
   doi:10.1051/0004-6361/202039061                                                  Yan, F., Casasayas-Barris, N., Molaverdikhani, K., et al. 2019, A&A, 632, A69
Gallet, F., Bolmont, E., Bouvier, J., et al. 2018, A&A, 619, A80                    Yan, F., Wyttenbach, A., Casasayas-Barris, N., et al. 2021, A&A, 645, A22

                                                                                                                                    Article number, page 11 of 18
You can also read